Identification of LPI Radar Signal Modulation using Bi-coherence Analysis and Artificial Neural Networks Techniques

This paper presents Higher Order Spectral Analysis (HOSA) and Artificial Neural Network techniques for identification of LPI (Low Probability of Intercept) Radar signal. Common Spectral analysis and conventional methods fail to detect low powered emissions of LPI Radars and even normal radars in noisy environments. This leads us to use Higher Order Spectral Analysis (HOSA) techniques (bi-spectrum, bi-coherence etc.,) enabling us to extract much more information from the same intercept and hence facilitating detection. Different types of radar waveforms used by LPI radars (e.g, pulse, LFM, phase coded using Barker code or Frank code etc.,) are simulated and then analyzed by bi-coherence analysis technique. An Artificial Neural Network (ANN) is trained on the results obtained by bi-coherence analysis, so that it will be able to detect and identify the LPI radar signal whose type is unknown when received. The results obtained clearly indicate the promising capability of the HOSA techniques to identify the type of LPI signal even with SNRs as low as –3 dB.